def test_leave_one_out(self):
correct = 0
k = 3
model = kNN.train(xs, ys, k)
predictions = [1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1]
for i in range(len(predictions)):
model = kNN.train(xs[:i] + xs[i + 1 :], ys[:i] + ys[i + 1 :], k)
prediction = kNN.classify(model, xs[i])
self.assertEqual(prediction, predictions[i])
if prediction == ys[i]:
correct += 1
self.assertEqual(correct, 13)
def test_leave_one_out(self):
correct = 0
k = 3
model = kNN.train(xs, ys, k)
predictions = [1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1]
for i in range(len(predictions)):
model = kNN.train(xs[:i] + xs[i + 1:], ys[:i] + ys[i + 1:], k)
prediction = kNN.classify(model, xs[i])
self.assertEqual(prediction, predictions[i])
if prediction == ys[i]:
correct += 1
self.assertEqual(correct, 13)
def test_classify(self):
k = 3
model = kNN.train(xs, ys, k)
result = kNN.classify(model, [6, -173.143442352])
self.assertEqual(result, 1)
result = kNN.classify(model, [309, -271.005880394])
self.assertEqual(result, 0)
def test_classify(self):
k = 3
model = kNN.train(xs, ys, k)
result = kNN.classify(model, [6, -173.143442352])
self.assertEqual(result, 1)
result = kNN.classify(model, [309, -271.005880394])
self.assertEqual(result, 0)
def train_model(self, training_ffts, training_labels):
"""
Takes a set of training examples + corresponding true labels and returns a
trained SOM and kNN.
"""
som_size = self.params['som_size']
som_iterations = self.params['som_iterations']
som_learning_rate = self.params['som_learning_rate']
knn_k = self.params['knn_k']
som = SelfOrganizingMap(size=som_size,
n_iterations=som_iterations,
learning_rate=som_learning_rate)
column_vectors = np.hstack(training_ffts).T
np.random.shuffle(column_vectors)
som.fit(column_vectors)
training_sequences = [
self.sound_fft_to_string(sound_fft, som)
for sound_fft in training_ffts
]
knn_model = kNN.train(training_sequences, training_labels, knn_k)
return som, knn_model
def test_calculate_model(self):
k = 3
model = kNN.train(xs, ys, k)
self.assertEqual(model.classes, set([0, 1]))
n = len(xs)
for i in range(n):
self.assertAlmostEqual(model.xs[i, 0], xs[i][0], 4)
self.assertAlmostEqual(model.xs[i, 1], xs[i][1], 4)
self.assertEqual(model.ys[i], ys[i])
self.assertEqual(model.k, k)
def test_calculate_model(self):
k = 3
model = kNN.train(xs, ys, k)
self.assertEqual(model.classes, set([0, 1]))
n = len(xs)
for i in range(n):
self.assertAlmostEqual(model.xs[i, 0], xs[i][0], places=4)
self.assertAlmostEqual(model.xs[i, 1], xs[i][1], places=4)
self.assertEqual(model.ys[i], ys[i])
self.assertEqual(model.k, k)
def test_model_accuracy(self):
correct = 0
k = 3
model = kNN.train(xs, ys, k)
predictions = [1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]
for i in range(len(predictions)):
prediction = kNN.classify(model, xs[i])
self.assertEqual(prediction, predictions[i])
if prediction == ys[i]:
correct += 1
self.assertEqual(correct, 15)
def knn(self):
xs = [[-53.0, -200.78], [117.0, -267.14], [57.0, -163.47],
[16.0, -190.3], [11.0, -220.94], [85.0, -193.94],
[16.0, -182.71], [15.0, -180.41], [-26.0, -181.73],
[58.0, -259.87], [126.0, -414.53], [191.0, -249.57],
[113.0, -265.28], [145.0, -312.99], [154.0, -213.83],
[147.0, -380.85], [93.0, -291.13]]
ys = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0]
k = 3
model = kNN.train(xs, ys, k)
return model
def test_model_accuracy(self):
correct = 0
k = 3
model = kNN.train(xs, ys, k)
predictions = [1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]
for i in range(len(predictions)):
prediction = kNN.classify(model, xs[i])
self.assertEqual(prediction, predictions[i])
if prediction == ys[i]:
correct += 1
self.assertEqual(correct, 15)
def test_calculate_probability(self):
k = 3
model = kNN.train(xs, ys, k)
weights = kNN.calculate(model, [6, -173.143442352])
self.assertAlmostEqual(weights[0], 0.0, 6)
self.assertAlmostEqual(weights[1], 3.0, 6)
weights = kNN.calculate(model, [309, -271.005880394])
self.assertAlmostEqual(weights[0], 3.0, 6)
self.assertAlmostEqual(weights[1], 0.0, 6)
weights = kNN.calculate(model, [117, -267.13999999999999])
self.assertAlmostEqual(weights[0], 2.0, 6)
self.assertAlmostEqual(weights[1], 1.0, 6)
def test_calculate_probability(self):
k = 3
model = kNN.train(xs, ys, k)
weights = kNN.calculate(model, [6, -173.143442352])
self.assertAlmostEqual(weights[0], 0.0, places=6)
self.assertAlmostEqual(weights[1], 3.0, places=6)
weights = kNN.calculate(model, [309, -271.005880394])
self.assertAlmostEqual(weights[0], 3.0, places=6)
self.assertAlmostEqual(weights[1], 0.0, places=6)
weights = kNN.calculate(model, [117, -267.13999999999999])
self.assertAlmostEqual(weights[0], 2.0, places=6)
self.assertAlmostEqual(weights[1], 1.0, places=6)
def train_model(self, training_ffts, training_labels):
"""
Takes a set of training examples + corresponding true labels and returns a
trained SOM and kNN.
"""
som_size = self.params['som_size']
som_iterations = self.params['som_iterations']
som_learning_rate = self.params['som_learning_rate']
knn_k = self.params['knn_k']
som = SelfOrganizingMap(size=som_size,
n_iterations=som_iterations,
learning_rate=som_learning_rate)
column_vectors = np.hstack(training_ffts).T
np.random.shuffle(column_vectors)
som.fit(column_vectors)
training_sequences = [self.sound_fft_to_string(sound_fft, som) for sound_fft in training_ffts]
knn_model = kNN.train(training_sequences, training_labels, knn_k)
return som, knn_model
def knn(self): xs = [[-53.0, -200.78], [117.0, -267.14], [57.0, -163.47], [16.0, -190.3], [11.0, -220.94], [85.0, -193.94], [16.0, -182.71], [15.0, -180.41], [-26.0, -181.73], [58.0, -259.87], [126.0, -414.53], [191.0, -249.57], [113.0, -265.28], [145.0, -312.99], [154.0, -213.83], [147.0, -380.85], [93.0, -291.13]] ys =[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0] k=3 model = kNN.train(xs,ys,k) return model
def knn(self):
xs = [[1.0, 1.0], [1.0, 2.0], [1.0, 3.0], [1.0, 4.0]]
ys = [1, 1, 0, 0]
k = 3
model = kNN.train(xs, ys, k)
return model
features = tfidf.toarray() #将tf-idf矩阵抽取出来，元素a[i][j]表示j词在i类文本中的tf-idf权重
print(features.shape)
target = [c for (d, c) in documents]
train_set1 = features[:1500, :]
target_train = target[:1500]
test_set1 = features[1500:, :]
target_test = target[1500:]
from sklearn.svm import SVC
svclf = SVC(kernel='linear', probability=True)
svclf.fit(train_set1, target_train)
pred_svc = svclf.predict(test_set1)
print('SVM=', sum(pred_svc == target_test) / len(target_test))
from Bio import kNN
from scipy import spatial
model = kNN.train(train_set1, target_train, 7)
dd = [
kNN.classify(model, t, distance_fn=spatial.distance.cosine)
for t in test_set1
]
print('KNN_cos=',
sum(np.array(dd) == np.array(target_test)) / len(target_test))
from sklearn.neighbors import KNeighborsClassifier
knnclf = KNeighborsClassifier(n_neighbors=7) #default with k=5
knnclf.fit(train_set1, target_train)
pred_knn = knnclf.predict(test_set1)
print('KNN_eu=', sum(pred_knn == target_test) / len(target_test))
def knn(self): xs = [[1.0, 1.0], [1.0, 2.0], [1.0, 3.0], [1.0, 4.0]] ys =[1, 1, 0, 0] k=3 model = kNN.train(xs,ys,k) return model
